检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Xiao, Zhiqing Wang, Haobo Jin, Ying Feng, Lei Chen, Gang Huang, Fei Zhao, Junbo College of Computer Science and Technology Zhejiang University China School of Software Technology Zhejiang University China Key Lab of Intelligent Computing based Big Data of Zhejiang Province Zhejiang University China CUHK-SenseTime Joint Lab The Chinese University of Hong Kong Hong Kong School of Computer Science and Engineering Nanyang Technological University Singapore Alibaba Group China

Unsupervised domain adaptation (UDA) is a pivotal form in machine learning to extend the in-domain model to the distinctive target domains where the data distributions differ. Most prior works focus on capturing the inter-domain transferability but largely overlook rich intra-domain structures, which empirically results in even worse discriminability. In this work, we introduce a novel graph SPectral Alignment (SPA) framework to tackle the tradeoff. The core of our method is briefly condensed as follows: (i)-by casting the DA problem to graph primitives, SPA composes a coarse graph alignment mechanism with a novel spectral regularizer towards aligning the domain graphs in eigenspaces;(ii)-we further develop a fine-grained message propagation module — upon a novel neighbor-aware self-training mechanism — in order for enhanced discriminability in the target domain. On standardized benchmarks, the extensive experiments of SPA demonstrate that its performance has surpassed the existing cutting-edge DA methods. Coupled with dense model analysis, we conclude that our approach indeed possesses superior efficacy, robustness, discriminability, and transferability. Code and data are available at: https://***/CrownX/SPA. Copyright © 2023, The Authors. All rights reserved.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Study of visual processing techniques for dynamic speckles: A comparative analysis

arXiv

引用

arXiv 2021年

作者： Chatterjee, Amit Dhanotiya, Jitendra Bhatia, Vimal Prakash, Shashi Signals and Software Group Discipline of Electrical Engineering Indian Institute of Technology Indore Indore453446 India Photonics laboratory Department of Electronics and Instrumentation Engineering Institute of Engineering and Technology Devi Ahilya University Indore452001 India

Dynamic activity of a speckle pattern is a useful non-destructive tool to study various activities (e.g. ripening, drying, osmosis, Brownian motion, etc.) in biological samples. For analysis of dynamic speckle pattern, visual methods are preferred over numerical one because of their ability to generate full field 2D activity map. Main visual techniques used to obtain information from speckle patterns are Fujii’s method, generalized difference (GD), weighted generalized difference, mean windowed difference (MWD), structural function (SF), modified structural function (MSF) etc. In this work, a comparative analysis of major visual techniques for natural gum sample collected from Vachellia nilotica tree is carried out. From obtained results it is observed that Fujii’s method presents negative activity map due to its weighting coefficient. Although this problem is circumvented in GD and MWD based techniques;they suffer from high time complexity problem. In order to improve computational efficiency, SF and MSF methods have been proposed. These have time complexity in the order of Fujii’s method along with a modified activity map. Activity parameters for all methods were estimated and tabulated for further comparison. Obtained results conclusively establish SF based method as an optimum tool for visual inspection of dynamic speckle data. © 2021, CC BY.

关键词： Speckle

来源：评论

学校读者我要写书评

暂无评论

Defect detection and NDE of low-modulus PMMA material using mechanical loading and WFT analysis

arXiv

引用

arXiv 2021年

作者： Dhanotia, Jitendra Chatterjee, Amit Bhatia, Vimal Prakash, Shashi Signals and Software Group Discipline of Electrical Engineering Indian Institute of Technology Indore Indore453446 India Photonics Laboratory Institute of Engineering & Technology Devi Ahilya University Khandwa Road Indore452001 India

The detection and analysis of defects in objects having complex shape is difficult using theoretical modeling. In present communication, we report our investigations undertaken towards detection of defects and non-destructive evaluation of low-modulus sample of Poly Methyl Metha Acrylate (PMMA) material using phase measuring deflectometry. Collimated light from He-Ne laser illuminates a sinusoidal grating. The resulting sinusoidal fringe pattern is projected onto the PMMA sample. The light reflected through the sample is recorded using Charge Couple Device (CCD) camera. When the sample was loaded along z-axis, the surface slope of the defective portions varies in an anomalous manner with respect to un-defective portions of the specimen. The prominent location and visualization of defects have been identified using windowed Fourier transform technique (WFT). The 3D phase plots reconstruct the profile of the specimen with presence of defects. The slope of the phase provides the information regarding defects and surface variation of the object. Experimental results conclusively establish the viability of the technique. © 2021, CC BY.

关键词： Defects

来源：评论

学校读者我要写书评

暂无评论

Correction to: The potential of an adaptive computerized dynamic assessment tutor in diagnosing and assessing learners’ listening comprehension

引用

Education and Information Technologies 2024年第14期29卷 19231-19231页

作者： Izadi, Mehri Izadi, Maliheh Heidari, Farrokhlagha Department of English Language and Literature University of Sistan and Baluchestan Zahedan Iran Software Engineering Research Group Delft University of Technology Delft The Netherlands

来源：评论

学校读者我要写书评

暂无评论

Cutting Force Similarity Calculation in Milling Process Using Siamese LSTM Structure

Cutting Force Similarity Calculation in Milling Process Usin...

引用

Mechatronics and Electrical engineering (MEEE), International Conference on

作者： Juheon Kwak Wonkeun Jo Soomin Lee Hyein Kim Jeongin Koo Dongil Kim Division of Artificial Intelligence and Software Ewha Womans University Seoul South Korea Computer Science and Engineering Chungnam National University Daejeon South Korea Manufacturing System R&BD Group Korea Institute of Industrial Technology Cheonan South Korea Department of Data Science Ewha Womans University Seoul South Korea

Cutting force is a key factor in machining processes. Cutting force similarity is required for several important issues, namely: stability evaluation, process control, and process parameter setting. This study employed a long short-term memory (LSTM) with Siamese architecture to measure the similarity of the cutting forces in a milling process. The Siamese LSTM was trained with time series data of the vertical cutting force collected from a cutting tool during the milling process to calculate the similarity. For evaluation, dynamic time warping (DTW), a common approach used to calculate the similarity of time series data, was employed for comparison with the Siamese LSTM. Experimental results showed that the proposed Siamese LSTM outperformed the conventional DTW-based similarity calculation.

关键词： Training Electrical engineering Mechatronics Force measurement Force Time series analysis Process control

来源：评论

学校读者我要写书评

暂无评论

Development of a Hypersonic Entry Test Platform

Development of a Hypersonic Entry Test Platform

引用

AIAA Science and technology Forum and Exposition, AIAA SciTech Forum 2025

作者： Johnson, Breanna J. Johnson, Kate E. Jang, Peter Steffes, Stephen R. Fritz, Matthew P. Hawke, Veronica Garcia, Joseph A. Morr, Donald E. NASA Johnson Space Center Flight Mechanics and Trajectory Design Branch HoustonTX77058 United States NASA Johnson Space Center Applied Aeroscience and CFD Branch HoustonTX77058 United States Aerodyne Industries Applied Aeroscience and CFD Branch HoustonTX77058 United States Charles Stark Draper Laboratory Navigation Algorithms and Software Space GNC Group HoustonTX77058 United States Science Technology Corporation Systems Analysis Office Mountain ViewCA United States NASA Ames Research Center NASA Systems Analysis Office Mountain ViewCA94035 United States Millennium Engineering and Integration Company ArlingtonVA22202 United States

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

NASA is designing and building a low-cost, scalable, rapidly developed test platform to lower the barrier to entry for entry, descent, and landing (EDL) technologies. This HyperSonic Testing for Entry Platform, HyperSTEP, will enable the low-cost flight demonstration of hypersonic entry technologies, leveraging a publishable government reference design to further enable future government and industry landers. The HyperSTEP vehicle leverages a lifting body shape that was originally designed for a NASA human-class Mars mission concept vehicle, the CobraMRV. Lifting body vehicles offer increased volume efficiency and greater range control capabilities over capsules, making them an attractive class of vehicles to further develop for small to large payload return missions from the Moon, Mars, and beyond. Progress has been made in the development of the HyperSTEP vehicle in the areas of aerodynamics, aerothermodynamics, structural design and packaging, propulsion, guidance navigation and control (GN&C), decelerators (i.e., parachutes), and the thermal protection system (TPS). This paper details progress made in refining the mission concept of operations (ConOps) for the first notional missions and the development of the aerodynamic and aerothermodynamic databases, GN&C, and parachute system, as the project moves into its build and test phase. © 2025, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： NASA

来源：评论

学校读者我要写书评

暂无评论

Smart Grid Cyber Attacks: Overview, Threats, And Countermeasures

Smart Grid Cyber Attacks: Overview, Threats, And Countermeas...

引用

International Conference on Intelligent Systems Applications to Power Systems (ISAP)

作者： Muhammad Adnan Khan Ahmed Mohammed Saleh Muhammad Waseem Vokony István Department of Electrical Engineering HITEC University Taxila Rawalpindi Pakistan Department of Electric Power Engineering Faculty of Electrical Engineering and Informatics Budapest University of Technology and Economics BME V1 Building Egry József u. 18 Budapest Hungary Department of Electronic Engineering International Renewable and Energy System Integration Research Group (IRESI) Maynooth University Co. Kildare Ireland Lero - Science Foundation Ireland Research Centre for Software School of Business Innovation Value Institute (IVI) Maynooth University Co. Kildare Ireland

ISBN: (数字)9798331531751

ISBN: (纸本)9798331531768

The smart grid paradigm brings about a significant transformation in conventional power grid networks through the seamless integration of cutting-edge computer, communication, and sensing technologies. This integration promotes optimal bidirectional power circulation, cultivating a digitally interconnected power grid that offers a multitude of advantages. The aforementioned benefits encompass improved inter-system communication, granularized control capabilities, bidirectional energy transmission, and the capacity to efficiently adapt to demand fluctuations. Notwithstanding its capacity for revolutionization, the smart grid presents a pivotal apprehension: cybersecurity. It is of the utmost importance to prioritize information security; however, the implementation of smart grid technology presents novel susceptibilities, specifically with regard to cyber threats. This article delves into the dynamic and changing environment of cybersecurity risks as they pertain to smart grid technology. Beginning with established methods of securing conventional power infrastructures, this section lays the groundwork for comprehending the ever-changing cybersecurity challenges of the future. Following this, an analysis is conducted on the progressively intricate cyber threats that the smart grid encounters. The article delineates a multitude of prevalent cyberattack vectors and suggests preventative measures to alleviate the associated hazards. By conducting an extensive investigation, this study seeks to illuminate the critical significance of cybersecurity in safeguarding the dependability and robustness of the smart grid infrastructure.

关键词： Vectors Robustness Smart grids Power system reliability Sensors Stakeholders Computer security Protection Cyberattack Resilience

来源：评论

学校读者我要写书评

暂无评论

A Graph Representation Learning-Based Method for Event Prediction

引用

IET Information Security 2025年第1期2025卷

作者： Xi Zeng Guangchun Luo Ke Qin Pengyi Zheng School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu Sichuan China *** Research Department The 30th Research Institute of China Electronics Technology Group Corporation Chengdu Sichuan China *** School of Software Engineering Xi’an Jiaotong University Xi’an Shaanxi China ***

With the continuous advancement of big data and artificial intelligence technologies, event prediction is increasingly being utilized across a multitude of domains. Predicting events allows for the exploration of the developmental trajectories and summarization of patterns associated with these events. However, events typically encompass a myriad of elements and intricate relationships, necessitating an enhancement in the precision of event prediction. However, the existing methods suffer from poor data quality, insufficient feature information, limited generalization capability of the models, and difficulties in evaluating prediction errors. This paper proposes a novel event prediction method based on graph representation learning, aiming to improve the accuracy of event prediction while reducing the time cost. By constructing causal graphs and introducing the script event simulation method, the architecture combines graph neural networks (GNNs) with BERT to simplify the event prediction process. Additionally, by combining GNNs with pretrained language models, a dynamic graph representation learning method is proposed. This means that a unified graph representation learning model can be built by following specific rules, thus predicting the development trajectory of events more accurately. The study evaluates the effectiveness of dynamic graph representation learning technology in a specific scenario, specifically in the context of employee career choices. By converting the career graph of employees into low-dimensional representations, the effectiveness of the dynamic graph representation learning method in predicting employee career decisions is validated. This innovation not only improves the accuracy of event prediction but also helps better understand and respond to complex event relationships in practical applications, providing decision-makers with more powerful information support. Therefore, this research has important theoretical and practical significanc

关键词： career choice prediction event prediction graph representation learning heterogeneous graph

来源：评论

学校读者我要写书评

暂无评论

Setar: A Novel Method of Stacking Ensemble Learning for Thai Sentiment Analysis Using Roberta and Hybrid Feature Representation

SSRN

引用

SSRN 2022年

作者： Thiengburanathum, Pree Charoenkwan, Phasit Department of Software Engineering College of Arts Media and Technology Chiang Mai University Chiang Mai50200 Thailand Department of Modern Management Information and Technology College of Arts Media and Technology Chiang Mai University Chiang Mai50200 Thailand Research Group of Modern Management Information and Technology College of Arts Media and Technology Chiang Mai University Chiang Mai50200 Thailand

Sentiment classification of social media posts is among the most challenging and time-consuming tasks for analysts. This is particularly true when applied to languages that employ scriptio continua, such as the Thai language, in which there are no spaces between written words and where there is no end of sentence punctuation. Thai is considered a scarce-resource language as few datasets are available to researchers. Although machine-learning (ML) and deep-learning (DL) algorithms can identify sentiment classification polarity in Thai, the performance of the existing classification models are still inadequate. This study proposes a novel stacking ensemble learning technique for identifying sentiment classification polarity in the Thai language, SETAR. Our stacking ensemble strategy utilized the pre-trained Thai language model (WangChanBERTa), based on a Robustly Optimized BERT Pretraining Approach (RoBERTa) architecture to form a feature vector. This feature was combined with three distinct feature vectors obtained from three well-known categories, namely Word2Vec, TF-IDF, and bag-of-words, as a new hybrid sentence representation. The base learners were trained using six chosen complex heterogeneous complex heterogeneous ML algorithms, including support vector machine (SVM), random forest (RF), extremely randomized trees (ET), light gradient boosting machine (LGBM), multi-layer perceptron (MLP), partial least squares (PLS), and logistic regression (LR) to enable the development of the final meta-learners. The results revealed that our proposed stacking ensemble model outperformed the baseline models of all classification metrics among the training and test sets, as was determined by extensive benchmarking, carried out on the four datasets, which included our developed sentiment corpus that domain experts annotated. © 2022, The Authors. All rights reserved.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Based Heart Rate Estimation Using Smart Shoes Sensor

Deep Learning Based Heart Rate Estimation Using Smart Shoes ...

引用

2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020

作者： Baek, Suwhan Eom, Heesang Hariyani, Yuli Sun Kim, Gwangho Roh, Jongryum Kim, Sayup Park, Chelsoo Kwangwoon University Department of Computer Engineering Seoul Korea Republic of Kwangwoon University Department of Software Seoul Korea Republic of Korea Institute of Industrial Technology Human Convergence Technology Group Ansan Korea Republic of

ISBN: (纸本)9781728161648

Although heart rate is an important biomarker of the physical condition at active states of users, it is still difficult to be measured due to an ambient noise and movements. There have been several approaches proposed to obtain stable measurements in an active condition. However, these methods still need direct contact to users, and thus additional equipment to keep the contact are requested, resulting in inconvenience of its usage. This paper proposes a method to estimate the heart rate of the user using activity information from smart shoes sensors, which is relatively easy and robust to be recorded. For the accurate estimation of the heart rate, a new design of deep neural networks is proposed. The architecture extracts features of time-sequential patterns of sensor data with implementing CNN and LSTM model together. The model was validated with a 'Leave-OneOut Cross-Validation method'. The results of the experiments are 10.21 ± 3.31 RMSE, 8.31 ± 2.81 MAE and 0.91 ± 0.09 correlation coefficient (Pearson) for the estimation of heart rate from smart shoes sensor data. © 2020 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：